slowly creeping along
Feb. 15th, 2009 12:32 am![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
kareinaI've been spending the past couple of days crunching numbers. "But, wait, weren't you supposed be done with number crunching by now?" you ask. Yes, indeed, but these aren't new numbers, really. You see, once upon a time, long ago now, my advisor instructed me to compile a file containing the compositions of one each muscovite, biotite, feldspar, and garnet-rim from each of my samples (choosing ones which are in close proximity to one another, and which look to be in equilibrium). He took said file, ran the data through the program Thermocalc to determine the "average pressure and temperature" for each sample (which the program finds by coming up with a set of chemical reactions that could produce those minerals and solving thermodynamic equations using those mineral compositions), and was pleased to note that all of the samples in a region gave very compatible answers. He then showed me (in a very quick display) how to use the program and gave me a folder of the necessary files. I tried using the program, and it crashed. So I set that task aside on the "to do later" list, and wrote the chapter with the numbers from his first pass of the program, highlighted in lurid colours to remind me that these, while likely close, aren't quite the final numbers.
It is now later, so I went back to him on Monday, got a new copy of the program, and now have it working smoothly. The program works in two steps. Step one you give it a guess as to the temperature and pressure at which the minerals form, feed it a text file with the compositions of the minerals (measured with the microprobe) and take the resultant text file it spits out, edit it slightly, and feed it to the other half of the program, which does the calculations and spits out a new text file with the results. One can then refine the results by taking the Pressure and Temperature it calculates and entering that in step one, and repeating the process. In addition, there is value in checking more than one set of minerals in a sample, and running each spot analysed on a given mineral. In theory, all of these runs for a single sample will give the same result (within error). If it doesn't, then it is necessary to figure out why, and which one is the "correct" answer. Sometimes one gets very different compositions for different spots on a single mineral because one of the analyses points was too near a crack in the grain, or there was an inclusion of another mineral close enough that the spot was contaminated.
Each set of calculations takes about two minutes, to create the text file, run step one, edit the new text file, run step to, and then copy the relevant parts of the third text file into a spreadsheet. However, there are often 20 to 60 sets of analyses that can be reasonably run for a sample, which, at two minutes each, adds up. This is why I'm not done yet. It is also a rather tedious process, though it is fun watching the results come in, one at a time, and watching where they plot on a graph. I'm starting to get an idea of whcih direction the temperature/pressure will shift based on the changes in composition. Fortunately, I am getting results which are reasonably consistent for any given sample, and they are all very similar to the numbers my advisor got for them all those months ago. If I can just keep at it, I might be able to finish this before going in to meet with him on Monday. Wish me luck.
In other news, I've been having fun playing with my hammer dulcimer, and wrote some words in praise of metamorphic rocks to help me remember a tune. Last night I had much fun dancing to the music of Xenos at the folk federation dance.
It is now later, so I went back to him on Monday, got a new copy of the program, and now have it working smoothly. The program works in two steps. Step one you give it a guess as to the temperature and pressure at which the minerals form, feed it a text file with the compositions of the minerals (measured with the microprobe) and take the resultant text file it spits out, edit it slightly, and feed it to the other half of the program, which does the calculations and spits out a new text file with the results. One can then refine the results by taking the Pressure and Temperature it calculates and entering that in step one, and repeating the process. In addition, there is value in checking more than one set of minerals in a sample, and running each spot analysed on a given mineral. In theory, all of these runs for a single sample will give the same result (within error). If it doesn't, then it is necessary to figure out why, and which one is the "correct" answer. Sometimes one gets very different compositions for different spots on a single mineral because one of the analyses points was too near a crack in the grain, or there was an inclusion of another mineral close enough that the spot was contaminated.
Each set of calculations takes about two minutes, to create the text file, run step one, edit the new text file, run step to, and then copy the relevant parts of the third text file into a spreadsheet. However, there are often 20 to 60 sets of analyses that can be reasonably run for a sample, which, at two minutes each, adds up. This is why I'm not done yet. It is also a rather tedious process, though it is fun watching the results come in, one at a time, and watching where they plot on a graph. I'm starting to get an idea of whcih direction the temperature/pressure will shift based on the changes in composition. Fortunately, I am getting results which are reasonably consistent for any given sample, and they are all very similar to the numbers my advisor got for them all those months ago. If I can just keep at it, I might be able to finish this before going in to meet with him on Monday. Wish me luck.
In other news, I've been having fun playing with my hammer dulcimer, and wrote some words in praise of metamorphic rocks to help me remember a tune. Last night I had much fun dancing to the music of Xenos at the folk federation dance.
